Nowadays there is often a need to build massively scalable distributed systems such as, for example: distributed databases that support petabytes of data and terabits per second read and write rates, cloud computing systems that balance load between geographical locations and server hardware, and content delivery networks that automatically find a suitable delivery point. Publication U.S. Pat. No. 7,978,631 describes a method, an apparatus, and a system for providing a data transfer network using an identification and communication mechanism. A data item that is a part of or related to an entity that is wanted to be accessible via the data transfer network is used to generate a unique identifier using a hash function. The accessible entity can be, for example, a data file, a data element of a hierarchical data structure, a process to be controlled via the data transfer network, or a device to be controlled via the data transfer network. The above-mentioned data item can comprise, for example, a name of a data file, a combination of a name of a hierarchical data structure and a name of an element of the data structure, identifier of a process or a patch job, and/or a name of a network element or a device connected to a network element. An advertisement message that specifies at least one address associated with the accessible entity is provided with the unique identifier. The advertisement message is broadcasted to the network elements of the data transfer network. A protocol data unit “PDU”, e.g. a data packet or a data frame, is addressed to the accessible entity in the data transfer network by retrieving, with the aid of the unique identifier, an address associated with accessible entity. Then, the protocol data unit is routed to the address associated with the accessible entity using an Open System Interconnection “OSI” Level 2 data link layer protocol, e.g. Ethernet, and/or a higher level protocol, e.g. an OSI Level 3 network layer protocol, e.g. Internet Protocol “IP”. The associations between the unique identifiers of the accessible entities and the addresses related to the accessible entities provide content based addressability which e.g. facilitates arranging scalable distributed databases, but maintaining these associations requires resources.